Comparison of the Reactivities of Dimethylsilylene (SiMe2 ... · 2) in Solution by Laser Flash Photolysis Methods Andrey G. Moiseev and William J. Leigh* Department of Chemistry,

Comparison of the Reactivities of Dimethylsilylene (SiMe2) and

Diphenylsilylene (SiPh2) in Solution by Laser Flash Photolysis

Methods

Andrey G. Moiseev and William J. Leigh*

Department of Chemistry, McMaster UniVersity, 1280 Main Street West,Hamilton, Ontario, Canada L8S 4M1

ReceiVed July 1, 2007

Laser flash photolysis studies have been carried out with the goal of comparing the reactivities ofdimethyl- and diphenylsilylene (SiMe2 and SiPh2, respectively) toward a comprehensive series ofrepresentative substrates in hexane solution under common conditions. The silylenes are generated usingdodecamethylcyclohexasilane (1) and 1,1,3,3-tetramethyl-2,2-diphenyl-1,2,3-trisilacyclohexane (9), re-spectively, as photochemical precursors. The reactions examined include O-H, O-Si, and M-H (M )Si, Ge, Sn) bond insertions, Lewis acid-base complexation with tetrahydrofuran, (1+2) cycloadditionto CdC and CtC multiple bonds, chlorine atom abstraction from CCl4, and reaction with molecularoxygen; the kinetic data for SiPh2 are accompanied by product studies in most cases. Further insight intothe mechanisms of these reactions is provided by the identification of reaction intermediates and/or transientproducts in some cases; notable examples include the reactions of SiPh2 with methoxytrimethylsilane,carbon tetrachloride, and molecular oxygen. With several of the substrates that were studied, comparisonof the kinetic data for SiPh2 to previously reported rate constants for reaction of dimesitylsilylene incyclohexane allows an assessment of the role of steric effects in affecting silylene reactivity. Rate constantscould also be determined for quenching of tetramethyldisilene (Si2Me4) with molecular oxygen, CCl4,and several other reagents that were studied.

Introduction

Despite great experimental and theoretical interest over thepast several decades in the chemistry of divalent organosiliconcompounds (silylenes),1 relatively little is known of their reactionkinetics in solution. The parent molecule (SiH2) and severalsimple substituted derivatives SiXY (X,Y) H, Me, or halogen)have received extensive attention in gas-phase kinetic studies,2-5

and the results of this work have served to provide a usefullink between theoretical and experimental studies of silylenereactivity in the gas phase. Until very recently however,6 onlydimethyl- and dimesitylsilylene (SiMe2 and SiMes2, respec-tively) had been successfully detected and studied in solutionby time-resolved spectroscopic methods,7-11 using the well-known oligosilane derivatives 1 and 2 as photochemical

precursors to the silylenes of interest. The first attempts to study

reactive silylenes by laser flash photolysis methods in solution

employed the acyclic trisilanes 312,13 and 414 as precursors to

the phenylated silylenes SiMePh and SiPh2, respectively, as it

was well known that photolysis of both compounds (as well as

1 and 2 and a number of other aryltrisilane derivatives) in low-

temperature glasses afforded the corresponding silylenes cleanly,

allowing their UV/vis absorption spectra to be characterized.15

These compounds were also well known to afford the corre-

sponding silylenes in high yields upon photolysis in solution at

room temperature. Unfortunately, they also undergo competing

photorearrangements under the latter conditions, producing

transient silene derivatives (e.g., 5; eq 1) whose strong UV/vis

absorptions occur in a similar spectral range as the more weakly

absorbing silylenes, obscuring the transient spectra to the point

where the silylenes of primary interest were missed altogether.* Corresponding author. E-mail: [email protected].(1) (a) Gaspar, P. P. In ReactiVe Intermediates, Vol. 3; Jones, M., Jr.,

Moss, R. A., Eds.; John Wiley & Sons: New York, 1985; pp 333-427;Vol. 2, 1981; pp 335-385; Vol. 1, 1978; pp 229-277. (b) Gaspar, P. P.;West, R. In The Chemistry of Organic Silicon Compounds, Vol. 2;Rappoport, Z., Apeloig, Y., Eds.; John Wiley and Sons: New York, 1998;pp 2463-2568. (c) Tokitoh, N.; Ando, W. In ReactiVe IntermediateChemistry; Moss, R. A., Platz, M. S., Jones, M., Jr., Eds.; John Wiley &Sons: New York, 2004; pp 651-715.(2) Safarik, I.; Sandhu, V.; Lown, E. M.; Strausz, O. P.; Bell, T. N. Res.

Chem. Intermed. 1990, 14, 105.(3) Jasinski, J. M.; Becerra, R.; Walsh, R. Chem. ReV. 1995, 95, 1203.(4) Becerra, R.; Walsh, R. Res. Chem. Kinet. 1995, 3, 263.(5) Becerra, R.; Walsh, R. Phys. Chem. Chem. Phys. 2007, 9, 2817.(6) Moiseev, A. G.; Leigh, W. J. J. Am. Chem. Soc. 2006, 128, 14442.(7) Levin, G.; Das, P. K.; Lee, C. L. Organometallics 1988, 7, 1231.(8) Levin, G.; Das, P. K.; Bilgrien, C.; Lee, C. L.Organometallics 1989,

8, 1206.(9) Shizuka, H.; Tanaka, H.; Tonokura, K.; Murata, K.; Hiratsuka, H.;

Ohshita, J.; Ishikawa, M. Chem. Phys. Lett. 1988, 143, 225.(10) Yamaji, M.; Hamanishi, K.; Takahashi, T.; Shizuka, H. J. Photo-

chem. Photobiol. A: Chem. 1994, 81, 1.

(11) Conlin, R. T.; Netto-Ferreira, J. C.; Zhang, S.; Scaiano, J. C.Organometallics 1990, 9, 1332.(12) Gaspar, P. P.; Boo, B. H.; Chari, S.; Ghosh, A. K.; Holten, D.;

Kirmaier, C.; Konieczny, S. Chem. Phys. Lett. 1984, 105, 153.(13) Gaspar, P. P.; Holten, D.; Konieczny, S.; Corey, J. Y. Acc. Chem.

Res. 1987, 20, 329.(14) Konieczny, S.; Jacobs, S. J.; Braddock Wilking, J. K.; Gaspar, P.

P. J. Organomet. Chem. 1988, 341, C17.(15) Michalczyk, M. J.; Fink, M. J.; De Young, D. J.; Carlson, C. W.;

Welsh, K. M.; West, R.; Michl, J. Silicon, Germanium, Tin Lead Compd.1986, 9, 75.

6277Organometallics 2007, 26, 6277-6289

10.1021/om700659d CCC: $37.00 © 2007 American Chemical SocietyPublication on Web 11/08/2007

As outlined in the preceding paper,16 we have recentlydeveloped two photochemical precursors to SiPh2 with the ideaof exploiting ring-strain effects to suppress the undesiredphotorearrangements characteristic of 2-phenyl-substituted trisi-lanes, which occur at the expense of silylene extrusion. Indeed,preliminary product and laser photolysis studies demonstratedthe approach to be successful. The first compound studied,trisilacycloheptane 6, affords SiPh2 and the accompanyingdisilacyclohexane derivative (7) in significantly higher yieldsthan 4 (eq 2), allowing the UV/vis spectrum of SiPh2 to berecorded in fluid solution and its reaction kinetics studied inspite of still-significant interference from transient absorptionsassignable to the silene coproduct, 8.6 Even better results wereobtained with the trisilacyclohexane derivative 9, which affordsSiPh2 and disilacyclopentane 10 in close to quantitative yieldsunder the same conditions (eq 3).16 This was shown by theresults of steady-state photolysis experiments with methanol(MeOH) and acetone, which trap the silylene as the formal O-Hinsertion and ene-addition products 12 and 13, respectively.16

Though the isomeric silene (11) still appears to be formed, itsyield is so low that it presents minimal interference in laserflash photolysis studies with the molecule. The UV/vis spectrumof SiPh2 thus emerges in almost identical form to that recordedin earlier matrix experiments with 4.15 In hexane solution inthe absence of scavengers, SiPh2 dimerizes to afford tetraphe-nyldisilene (Si2Ph4), whose spectrum was also recorded. Inter-estingly, it coincides almost exactly with that of the minor silenecoproduct; the situation thus remains complicated, but isnevertheless tractable. It is now possible to study the chemistryof this benchmark transient silylene in solution directly and atan unprecedented level of detail.

The first such study is the subject of the present paper: apreliminary survey of the kinetics of a broad selection of well-known silylene reactions in solution, focusing on SiPh2 and itsmethylated counterpart, SiMe2, with data recorded under acommon set of conditions so as to enable direct comparisonsbetween the two silylene derivatives. The specific reactionsstudied are shown in Chart 1 and include Lewis acid-base

complexation with THF, O-H, O-Si, and group 14 M-H (M) Si, Ge, Sn) !-bond insertions, reaction with molecular oxygen,(2+1) cycloaddition to alkenes, dienes, and alkynes, and chlorineatom abstraction from CCl4. Advantage is taken of the fact thatthe presence of the phenyl chromophore in SiPh2 results in ared-shift of the absorption spectra of the silylene and its reactionproducts relative to those of SiMe2, enhancing our ability todetect and characterize intermediates that may be involved in agiven silylene-substrate reaction. In several cases, this providesimportant new insight into the mechanisms of these reactions.

Results and Discussion

Laser flash photolysis experiments were carried out asdescribed in the preceding paper,16 using rapidly flowedsolutions of 9 (!10-4 M) and 1 (ca. 5 " 10-4 M) in dry,deoxygenated hexane and the pulses from a KrF excimer laser(248 nm, ca. 25 ns, ca. 100 mJ) for excitation. Silylene decayswere monitored at 530 nm for SiPh2 and 470 nm for SiMe2;they generally fit acceptably to first-order decay kinetics in theabsence of added scavenger, with lifetimes that varied over theranges 0.6-1.5 µs and 250-600 ns, respectively. As detailedin the preceding paper,16 the decay of the silylenes wasaccompanied by the growth of absorptions at shorter wave-lengths, due to the formation of the corresponding disilenesSi2Ph4 ("max ) 290, 370, 460 nm) and Si2Me4 ("max ) 360 nm)by silylene dimerization. The variation in the lifetimes of thesilylenes is due to an acute sensitivity to the presence of moisturein the solvent and the glassware, which was difficult to control(particularly on humid days) despite the routine of flame-dryingthe apparatus prior to each experiment. As expected, the initialintensities of the disilene signals also varied from experimentto experiment, correlating inversely with the silylene lifetimes.Addition of the substrates shown in Chart 1 caused the

pseudo-first-order decay of the silylenes to accelerate inproportion to the substrate concentration; in all cases but THF,this was accompanied by increases in the growth rates of thedisilene absorptions and reductions in their maximum intensities.Rate constants were determined by linear least-squares analysisof plots of the pseudo-first-order decay rate constants for decay(kdecay) of SiPh2 and SiMe2 according to eq 6, where k0 is thepseudo-first-order rate constant for decay of the species in theabsence of the substrate (Q) and kQ is the second-order rateconstant for the reaction with Q; these plots were linear in allbut one case (Vide infra). The data are summarized in Table 1,along with reported rate constants for reaction of seven of thesubstrates with SiMes2 in cyclohexane or hexane solution.11,16

Each experiment was completed by recording transient spectrain the presence of the scavenger at a concentration wherethe silylene lifetime was reduced to ca. 300 ns or less, and ishence determined mainly by reaction with the substrate. Thiswas done in an effort to probe for the formation of the primaryproduct(s) of the silylene-trapping reaction. These additionaldetails will be presented later in the paper.

The outcomes of these reactions are all well documented forSiMe2 and/or other transient silylenes,1,17 but only those withEt3SiH, dienes, alkenes, and acetone have been studied previ-ously for SiPh2.6,16,18-20 We therefore carried out NMR-scalephotolyses of ca. 0.05 M solutions of 9 in C6D12 containing ca.

(16) Moiseev, A. G.; Leigh, W. J. Organometallics 2007, 26, 6268-6276.

(17) Kira, M.; Iwamoto, T.; Ishida, S. Bull. Chem. Soc. Jpn. 2007, 80,258.(18) Bobbitt, K. L.; Gaspar, P. P. J. Organomet. Chem. 1995, 499, 17.

kdecay ) k0 + kQ[Q] (6)

6278 Organometallics, Vol. 26, No. 25, 2007 MoiseeV and Leigh

0.20 M AcOH, MeOTMS, Et3GeH, Bu3SnH, BTMSE, DMB,and CCl4, following the course of the photolyses by 1H NMRspectroscopy and GC/MS, the latter being recorded before andafter photolysis to 50-60% conversion of 9. The productsformed in these experiments (see Chart 1) were in most casesidentified by comparison of the NMR and GC/MS data for thecrude reaction mixtures to the corresponding data for authenticsamples or to literature data; only two of them (vinylsilirane23 and silirene 26) were tentatively identified by comparisonof the spectral data to reported data for closely relatedcompounds. As in the MeOH- and acetone-trapping experimentsdescribed in the preceding paper,16 yields were determinedrelative to consumed 9 from the slopes of concentration versustime plots constructed from the NMR data over the 0-25%conversion range. The photolysis mixtures were generally quiteclean, with the yield of disilacyclopentane 10 matching that ofthe SiPh2-derived product(s) reasonably closely in most cases.The chemical yields of the SiPh2-derived products identified inthese experiments are shown in Table 1 along with the kineticdata. Additional specific details pertaining to the steady state

and laser flash photolysis experiments with each of the substratesstudied are presented below.

Finally, for several of the reagents studied with the SiMe2precursor 1 (AcOH, CCl4, TBE, acetone, DMP, isoprene, O2,and BTMSE), the reduction in the maximum transient absor-bance at 360 nm (due to Si2Me4) with increasing substrateconcentration was accompanied by a distinct shortening of thelifetime of the absorption over the substrate concentration rangesstudied, with the decays fitting well to first-order kinetics. Rateconstants for quenching of Si2Me4 were determined from plotsof kdecay versus [Q] according to eq 6 and are listed in Table 2.

Reactions of SiPh2 and SiMe2 with Oxygenated Substrates.

The Lewis acid-base complexation of silylenes with ethers andother heteroatom-containing substrates is well known and hasbeen studied extensively in low-temperature matrixes withSiMe2 and a number of other reactive silylene derivatives.21-25

(19) Sakurai, H.; Kobayashi, Y.; Nakadaira, Y. J. Organomet. Chem.1976, 120, C1.(20) Tortorelli, V. J.; Jones, M., Jr.; Wu, S.; Li, Z.Organometallics 1983,

2, 759.

(21) Gillette, G. R.; Noren, G. H.; West, R. Organometallics 1987, 6,2617.(22) Ando, W.; Sekiguchi, A.; Hagiwara, K.; Sakakibara, A.; Yoshida,

H. Organometallics 1988, 7, 558.(23) Gillette, G. R.; Noren, G. H.; West, R. Organometallics 1989, 8,

487.(24) Belzner, J.; Ihmels, H. AdV. Organomet. Chem. 1999, 43, 1.(25) Boganov, S. E.; Faustov, V. I.; Egorov, M. P.; Nefedov, O. M.Russ.

Chem. Bull., Int. Ed. 2004, 53, 960.

Chart 1

Table 1. Diphenylsilylene Trapping Products and Absolute Rate Constants for Reaction of Diphenyl-, Dimethyl-, andDimesitylsilylene with Various Silylene Scavengers in Hexane Solution at 25 °C

trapping agent product(s) kQSiPh2/109 M-1 s-1 kQSiMe2/109 M-1 s-1 kQSiMes2/109 M-1 s-1

MeOHa 12 (93%) 13.4 ( 1.0 18.1 ( 1.5 0.82 ( 0.03acetonea 13 (86%) 14.4 ( 2.0 14.1 ( 1.2 8.3 ( 0.4THF Ph2Si-THF complex 15.2 ( 1.3 17.3 ( 1.5MeOSiMe3 14 (54%) 4.2 ( 0.7 6.2 ( 0.6b < 0.001d

AcOH 16 (84%) 10.1 ( 1.0 15.9 ( 1.0O2 17 (R ) Ph)c 0.12 ( 0.02 0.47 ( 0.04 0.032 ( 0.004d

Et3SiH 19 3.3 ( 0.2a 3.6 ( 0.3 0.079 ( 0.004d

Et3GeH 20 (90%) 3.1 ( 0.2 2.5 ( 0.2n-Bu3SnH 21 (87%) 5.3 ( 0.6 18.9 ( 1.4cyclohexene 2218 7.9 ( 0.7 7.8 ( 1.0 0.0028 ( 0.0003d

DMP c 7.9 ( 1.6e 11.7 ( 0.5DMB 23 (87%) + 24 (<18%) 14.5 ( 1.3 15.9 ( 0.7 0.0088 ( 0.0007d

isoprene c 13.6 ( 1.4 19.8 ( 0.7BTMSE 26 (67%) 7.6 ( 0.6 13.8 ( 0.5TBE c 8.3 ( 0.7e 17.8 ( 0.6CCl4 27 (78%) 1.37 ( 0.09 3.4 ( 0.2

a Data from ref 16. bRate constant obtained from linear least-squares analysis of kdecay data at low concentrations (0.1-0.5 mM). cProduct not determined.d In cyclohexane at 25 °C; data from ref 11. eData from ref 6.

Comparison of the ReactiVities of SiMe2 and SiPh2 in Solution Organometallics, Vol. 26, No. 25, 2007 6279

Kinetic studies have been limited to SiH2 and SiMe2 in the gasphase26,27 and SiMe2 in solution.8,10 Though the complexationprocess is ultimately nonproductive in the cases of unstraineddialkyl ethers,28 it has the effect of moderating the reactivity ofsilylenes toward other substrates in ether compared to hydro-carbon solvents8,24,29,30 and forms the basis for our currentunderstanding the mechanisms of the reactions of silylenes withalcohols, alkoxysilanes, ketones, and strained cyclic ethers suchas oxetanes and oxiranes.1

Lewis acid-base complexation of SiPh2 with THF proceedsat close to the diffusion limit in hexane solution and leads tothe formation of new, longer-lived absorptions at shorterwavelengths ("max # 350 nm), which appear to grow in over asimilar time scale as the silylene decay over the 0.1-0.5 mMconcentration range in THF (see Figure 1a). The spectrum ofthe species is similar to that reported for the 2-methyltetrahy-drofuran complex of SiMes2 ("max # 350 nm) at 77 K21 and isred-shifted compared to that of the complex of SiMe2 ("max #310 nm) with THF in cyclohexane;8,10 it can thus be assignedwith confidence to the SiPh2-THF Lewis acid-base complex.A similar trend in "max is known for the complexes of thegermanium analogues of the three silylenes with THF.31 Disileneformation is slowed quite substantially over this concentrationrange, but appears to proceed in similar yield to that in purehexane solution. The behavior is illustrated in Figure 1a, whichshows transient absorption spectra recorded with a solution of9 in hexane containing 0.3 mM THF at four time intervals afterthe laser pulse and transient growth/decay profiles at 530, 460,and 350 nm. A dotted line has been placed at the 500 ns pointof the growth/decay profiles to emphasize the equivalence ofthe silylene decay rate and the growth rate of the complex, whichare both just discernible on the time scale employed for thisparticular experiment. The absorption intensity at 460 nm growsto a maximum over the first 5 µs after the pulse, 4-5 timesslower than in the absence of THF, and then decays at the samerate as the absorption due to the complex.The long-lived residual absorption present in the 530 nm

decay profile is due to silene 11, which absorbs very weakly at

this wavelength, and not to residual free SiPh2 in equilibriumwith the complex. This conclusion is supported by the fact thatthe long-lived absorption remained of similar relative intensitywith solutions of 9 containing up to 0.05 M THF, where thelifetime of the free silylene was too short to be detected, and isconsistent with an equilibrium constant in excess of 105 M-1

for formation of the silylene-ether complex under theseconditions. On the other hand, the temporal profiles of theabsorptions at 350 and 460 nm varied only modestly over thisconcentration range. This indicates that disilene formation occursmainly via a mechanism involving dimerization of the complexunder these conditions, even in the presence of as little as 0.3mM THF. The lengthening of both the growth and decay timesof the disilene absorption in the presence of the ether comparedto those in its absence indicates that this process occurssignificantly more slowly than that involving the free silylene,which is close to diffusion-controlled.16 Analysis of the 350 nmsignal as a second-order decay afforded a rate coefficient of2k/! # 106 cm s-1.

Addition of dioxane to hexane solutions of 9 led to similareffects; the ether quenched the lifetime of SiPh2 at the diffusion-controlled rate (kQ ) (1.8 ( 0.1) " 1010 M-1 s-1) and led tothe formation of a new species exhibiting "max ) 370 nm. Thespecies decayed with a lifetime # # 5 µs in the presence of 11mM dioxane, with the concomitant growth of the disileneabsorption at 460 nm. These results are relevant to theinterpretation of the various steady-state photolysis experimentsreported throughout this paper, in which dioxane was employedas an internal integration standard. It clearly also serves as amoderator in the silylene-trapping reactions, though it appearsto have no effect on their ultimate outcomes.

Laser photolysis of hexane solutions of 1 in the presence ofTHF afforded quite similar results to those described previouslyin cyclohexane solution8,9 and allowed the determination of kTHF) 1.73 " 1010 M-1 s-1 for formation of the SiMe2-THFcomplex ("max ) 310 nm) in hexane at 25 °C. The rate constantis the same within experimental error as that for formation ofthe SiPh2-THF complex under the same conditions.The behavior observed in the presence of the reactive ether

methoxytrimethylsilane (MeOTMS) was markedly differentfrom that observed with THF. As with THF, addition of thealkoxysilane to hexane solutions of 9 led to reductions in thelifetime of SiPh2 over the 0-1 mM concentration range,affording the second-order rate constant kMeOTMS ) 4.2 " 109

M-1 s-1 from the resulting linear plot of kdecay versus [MeOT-MS]. Growth/decay profiles recorded at 460 nm showed theformation of Si2Ph4 to be strongly quenched over the 0.1-4.0mM concentration range in added MeOTMS, which is charac-teristic of essentially irreversible scavenging of SiPh2, theimmediate precursor to the disilene. At 350 nm, the characteristicgrowth/decay associated with the S0 f S2 transition of Si2Ph4in pure hexane16 was replaced with a stronger, more rapidlydecaying signal whose maximum intensity increased withincreasing alkoxysilane concentration, superimposed on alonger-lived decay which decreased in overall intensity withincreasing concentration. No further changes in either compo-nent of the 350 nm signal were observed upon increasing thealkoxysilane concentration above 4 mM. The lifetime of therapidly decaying component of the 350 nm signal was deter-mined by two-exponential decay analysis of traces recorded inthe presence of 1, 2, and 4 mM MeOTMS and was found to beindependent of concentration within this range (# ) 730 ( 5ns). The only unusual feature noted in these experiments wasthat the initial intensity of the silylene signal (at 530 nm)

(26) (a) Baggott, J. E.; Blitz, M. A.; Frey, H. M.; Lightfoot, P. D.; Walsh,R. Int. J. Chem. Kinet. 1992, 24, 127. (b) Becerra, R.; Cannady, J. P.; Walsh,R. J. Phys. Chem. A 1999, 103, 4457. (c) Becerra, R.; Cannady, J. P.; Walsh,R. Phys. Chem. Chem. Phys. 2001, 3, 2343. (d) Alexander, U. N.; King, K.D.; Lawrance, W. D. Phys. Chem. Chem. Phys. 2001, 3, 3085. (e) Alexander,U. N.; King, K. D.; Lawrance, W. D. J. Phys. Chem. A 2002, 106, 973. (f)Becerra, R.; Cannady, J. P.; Walsh, R. J. Phys. Chem. A 2003, 107, 11049.(g) Becerra, R.; Bowes, S. J.; Ogden, J. S.; Cannady, J. P.; Adamovic, I.;Gordon, M. S.; Almond, M. J.; Walsh, R. Phys. Chem. Chem. Phys. 2005,7, 2900.(27) Alexander, U. N.; King, K. D.; Lawrance, W. D.Phys. Chem. Chem.

Phys. 2001, 3, 3085.(28) Gu, T.-Y. G.; Weber, W. P. J. Am. Chem. Soc. 1980, 102, 1641.(29) Steele, K. P.; Weber, W. P. J. Am. Chem. Soc. 1980, 102, 6095.(30) Steele, K. P.; Weber, W. P. Inorg. Chem. 1981, 20, 1302.(31) Leigh, W. J.; Lollmahomed, F.; Harrington, C. R.; McDonald, J.

M. Organometallics 2006, 25, 5424.

Table 2. Rate Constants for Quenching of Si2Me4 byVarious Substrates in Hexane Solution at 25 °C, Determined

from Plots of kdecay vs [Q] According to Eq 6

reagent kQ/109 M-1 s-1

AcOH 2.7 ( 0.2CCl4 2.3 ( 0.2O2 0.29 ( 0.04b

DMP 0.5 ( 0.2b

isoprene 0.4 ( 0.2TBE 0.81 ( 0.04BTMSE 0.59 ( 0.05

a In most cases, the plots were constructed using data recorded at 5-9substrate concentrations; errors are reported as ( 2!. bFrom kdecay valuesrecorded at only two concentrations.


appeared to be reduced in the presence of MeOTMS, to an extentthat increased systematically with increasing concentration. Wedo not yet understand the reasons for this behavior, but can atleast rule out excited-state quenching on the basis of the productstudies described below.

Figure 1b shows transient absorption spectra recorded overvarious time windows after the laser pulse with a hexane solutionof 9 containing 4 mM MeOTMS, along with decay profilesrecorded at 350 and 480 nm. The decay profiles show that thelong-lived component of the 350 nm signal is in fact associatedwith the minor transient coproduct formed along with SiPh2 inthe photolysis pulse, which we assign to silene11.16 The lifetimeof the species is largely unaffected in the presence of thealkoxysilane, which is consistent with the silene assignment.32

The short-lived component can be assigned to the Lewis acid-base complex of SiPh2 with MeOTMS on the basis of itssimilarity to the spectrum of the SiPh2-THF complex (seeFigure 1a), while its first-order decay can be associated withthe rate of SiMe3 migration to yield the final product formed inthe reaction, alkoxydisilane 14 (eq 7). The identity of the finalproduct was confirmed by steady-state photolysis of 9 in C6D12containing 0.19 M MeOTMS, which afforded 10 and 14 inyields of 74% and 46%, respectively. A third product was alsodetected by GC/MS of the crude photolysate after ca. 60%conversion of 9 and was tentatively identified as trisilane 15,the (secondary) product of competing trapping of SiPh2 by 14.Its formation was in fact expected, as it is known thatalkoxydisilanes are considerably more effective silylene scav-engers than alkoxysilanes.33,34

The quenching of SiMe2 by MeOTMS was found to besomewhat different than the behavior observed with SiPh2.Addition of the alkoxysilane to hexane solutions of 1 resultedin the expected increase in the silylene decay rate constant, butthe resulting plot of kdecay versus [MeOTMS] exhibited down-ward curvature over the 0.1-5.0 mM concentration range. Anestimate of the limiting second-order rate constant (i.e., as[MeOTMS] f 0) of kQ ) 6.2 " 109 M-1 s-1 was thereforedetermined by linear least-squares analysis of data recorded overthe 0.1-0.4 mM concentration range. The value is ca. 3 timeslower than the rate constant for reaction of SiMe2 with MeOHunder the same conditions, in reasonable agreement with therate constant ratio determined by Steele et al. for reaction ofSiMe2 with MeOTMS and ethanol in dilute hydrocarbonsolution, which was based on competition experiments.34

Quenching of the silylene was accompanied by the appearanceof a new transient absorption centered at "max # 310 nm, whichcan be assigned to the SiMe2-MeOTMS complex on the basisof its similarity to the spectrum of the SiMe2-THF complex.8The absorption decayed with clean first-order kinetics and alifetime # ) 150 ns in hexane containing 5.3 mM MeOTMS,nearly the same as that exhibited by the much weaker signal at470 nm due to free SiMe2 at this concentration (# # 110 ns).The formation of Si2Me4, which is barely detectable in thepresence of 1.4 mM MeOTMS (see Supporting Information),could not be detected at the higher alkoxysilane concentration,and so we interpret the similar lifetimes at 470 and 350 nm asbeing due to the free silylene and the complex in mobileequilibrium and decaying predominantly via unimolecularrearrangement of the complex to yield the formal O-Si insertionproduct. The underlying reasons for the nonlinear dependenceof the silylene decay rate on alkoxysilane concentration are notclear at the present time, though we note that the behavior isconsistent with saturation kinetics, as would result if there werea pathway for reaction of the complex that involves a secondmolecule of alkoxysilane and regenerates SiMe2.

Our results are consistent with the two-step mechanismoriginally proposed by Weber and co-workers for the O-Siinsertion reaction with alkoxysilanes, involving initial Lewisacid complexation of silylene with the substrate,34 but addconsiderable new insight into the finer mechanistic details ofthe reaction. As with MeOH,16 formation of the complex is therate-determining step for silylene decay at low substrateconcentrations; however, this stage of the process is considerablyslower in the case of MeOTMS, as might be expected consider-

(32) Leigh, W. J.; Sluggett, G. W. Organometallics 1994, 13, 269.(33) Gu, T.-Y. G.; Weber, W. P. J. Organomet. Chem. 1980, 195, 29.(34) Steele, K. P.; Tzeng, D.; Weber, W. P. J. Organomet. Chem. 1982,

231, 291.

Figure 1. (a) Transient absorption spectra recorded 64-128 ns (-0-), 0.64-0.70 µs (-O-), 3.52-3.58 µs (-4-), and 17.76-17.92 µs (---)after the laser pulse, from laser flash photolysis of a 0.09 mM solution of 9 in deoxygenated hexane containing 0.3 mM THF. The insetshows transient growth/decay profiles recorded at 350, 460, and 530 nm; a dotted line has been drawn at!0.6 µs to illustrate the coincidencein the decay of the signal at 530 nm and the growth at 350 nm. (b) Transient absorption spectra recorded 0-64 ns (-0-), 1.01-1.07 µs(-O-), and 8.5-8.6 µs (-4-) after the pulse, from laser photolysis of 9 in hexane containing 4.0 mM MeOTMS; the inset shows transientdecay profiles recorded at 350 and 480 nm.


ing the lower basicity of the alkoxysilane compared to the etherand alcohol.35 Our results also show that the intermediatecomplex is not a steady-state intermediate, but rather constitutesa bottleneck in the reaction pathway owing to its relatively slowrate of conversion to the final product, 14. It is interesting tonote that Conlin and co-workers found the reaction of SiMes2with MeOTMS to be too slow to be detected under theconditions of their kinetic experiments with this silylene in fluidsolution and estimated an upper limit for the rate constant ofkMeOTMS < 106 M-1 s-1.11 This suggests that complexation withthe alkoxysilane is significantly more sensitive to steric effectsthan is the case for the reaction with MeOH, as might beexpected on the basis of the fact that complexation is the rate-determining step for silylene decay in both cases and consideringthe differences in steric bulk between the two substrates.

While the rate constants for formation of the silylene-methoxysilane complexes from SiPh2 and SiMe2 (see Table 1)are quite similar, larger differences are observed in the first-order rate constants for their decay. The decay process can mostreasonably be associated with the 1,2-SiMe3 migration that takesthe intermediate complex to the final product of the reaction.The ca. 5-fold slower decay rate constant observed for theSiPh2-MeOTMS complex (kdecay # 1.4 " 106 s-1) relative tothat of the SiMe2-MeOTMS complex under the same condi-tions (kdecay # 6.6 " 106 s-1) is consistent with a somewhathigher secondary barrier for formation of the net O-Si insertionproduct in the phenylated system. The rate constants for decayof the complexes are consistent with activation energies for thesecond step in the reaction on the order of Ea # 7.5 kcal mol-1,assuming pre-exponential factors on the order of 1012 s-1.However, it is clear from the unusual concentration dependenceobserved with SiMe2 that the reaction is significantly morecomplicated than this analysis might suggest, and further workwill be necessary in order to understand it completely.

The reactions of SiPh2 with acetone and acetic acid affordthe formal ene-type addition products 1316,18 and 16 (see eq 8),respectively, and both proceed with rate constants close to thediffusion-controlled limit. As might therefore be expected, therate constants are also quite similar to those measured forreaction of SiMe2 with these substrates. While it is reasonableto expect that these reactions also proceed via initial complex-ation, and indeed spectroscopic evidence for this has beenreported for the reaction of a nonenolizable ketone with SiMes2at cryogenic temperatures,36 we have been unable to detect anysigns of intermediate complexes in the reactions of these threesilylenes with acetone16 or AcOH in fluid solution (seeSupporting Information). We thus conclude that if such speciesare indeed generally involved in the reactions with these twosubstrates, they must be true steady-state intermediates, rear-ranging to the final products with rate constants large enoughthat the intermediates do not build up in high enough concentra-tions to enable detection (eq 9). This is consistent with the resultsof theoretical calculations for the reaction of SiH2 withacetone.26b

The reaction of transient silylenes with molecular oxygen hasbeen studied in low-temperature matrixes in the cases ofSiMes2,37 SiMe2,38 and SiMePh39 and has also been the subject

of theoretical calculations by several groups;37,39-41 kineticstudies of the reaction of the parent molecule (SiH2) with O2 inthe gas phase have also been reported.41 Ando and co-workersfirst studied the reaction of SiMes2 with O2 by UV/vis and IRspectroscopy in an oxygen matrix at 16 K.37 They reported theformation of a species exhibiting a weak UV/vis absorptionmaximum at "max # 320 nm, which they assigned to thecorresponding triplet silanone oxide (17; see eq 10) on the basisof IR spectroscopic evidence and RHF/6-31(g) calculations onthe triplet state of the parent silanone oxide, 3(H2Si-O2). Sanderand co-workers later studied the reactions of O2 with SiMe238

and SiMePh39 in 0.5% O2-doped argon matrixes by IR spec-troscopy and showed that the two species react rapidly at ca.40 K to form the corresponding dioxasiliranes (18), without theintervention of detectable intermediates. It is thus clear thatfollowing intersystem crossing to the singlet (ground) statesurface, the initially formed silanone oxide closes to the cor-responding dioxasilirane derivative via an extremely lowactivation barrier. This is supported by theoretical calculations,which place the height of the barrier at ca. 5 kcal mol-1 for theparent silanone oxide41 and lower still in substituted systemssuch as that derived from SiMePh.39

We have measured absolute rate constants for the reactionsof SiPh2 and SiMe2 with oxygen in hexane in the present study,obtaining values of kO2 ) 1.2 " 108 and 4.7 " 108 M-1 s-1,respectively, from the slopes of three-point plots of kdecay inargon-, air-, and O2-saturated hexane versus [O2]. The valuefor SiPh2 is only ca. 4 times higher than that reported by Conlinand co-workers for SiMes2 in cyclohexane solution,11 showingthat steric effects on the rate constant for the process are quitesmall, as might be expected. Interestingly, the reaction with O2is the slowest that we have examined with SiMe2 and SiPh2,which most likely reflects the fact that the reaction yields inthe first step a relatively high-energy triplet 1,3-biradical product.The present value for SiMe2 of kO2 ) 4.7" 108 M-1 s-1 revisesthe ca. 5-fold higher value reported by Levin et al., which wasbased on a single lifetime determination in air-saturated cyclo-hexane.8

Transient absorption spectra were recorded by laser photolysisof an O2-saturated hexane solution of 9, where the lifetime ofSiPh2 is reduced to # # 330 ns; the results of the experimentare shown in Figure 2. Transient decays recorded at 460 nmsuggest that disilene formation is largely quenched under theseconditions; the transient absorption observed at 460 nm appears(35) Pitt, C. G.; Bursey, M. M.; Chatfield, D. A. J. Chem. Soc., Perkin

Trans. 2 1976, 434.(36) Ando, W.; Hagiwara, K.; Sekiguchi, A. Organometallics 1987, 6,

2270.(37) Akasaka, T.; Nagase, S.; Yabe, A.; Ando, W. J. Am. Chem. Soc.

1988, 110, 6270.(38) Patyk, A.; Sander, W.; Gauss, J.; Cremer, D. Angew. Chem., Int.

Ed. Engl. 1989, 28, 898.

(39) Bornemann, H.; Sander, W. J. Am. Chem. Soc. 2000, 122, 6727.(40) Nagase, S.; Kudo, T.; Akasaka, T.; Ando, W. Chem. Phys. Lett.

1989, 163, 23.(41) Becerra, R.; Bowes, S.-J.; Ogden, J. S.; Cannady, J. P.; Adamovic,

I.; Gordon, M. S.; Almond, M. J.; Walsh, R. Phys. Chem. Chem. Phys.2005, 7, 2900.


to consist of only a single component, which is formed withthe laser pulse and decays to the prepulse level with clean first-order kinetics and a lifetime # # 1.6 µs. The much reducedlifetime of the transient absorption at this wavelength in O2-compared to Ar-saturated hexane provides additional supportfor its assignment to silene 11, as the estimated second-orderrate constant for its reaction with O2 (k # 4 " 107 M-1 s-1;calculated from the 1.6 µs lifetime and a value of [O2] ) 15mM42) is of the magnitude expected on the basis of reportedkinetic data for other silenes of this general structure.32 It isdifficult to rule out the possibility that the 460 nm signal inO2-saturated solution contains minor contributions from thesilylene dimer, Si2Ph4. However, if this is the case, then thefact that the signal decays with clean first-order kinetics suggeststhat Si2Ph4 and 11 have similar lifetimes and hence similar rateconstants for reaction with O2. This is, in fact, consistent withreported data for other phenyl-substituted disilenes.43

In addition to the prompt absorptions due to SiPh2 (# # 330ns) and silene 11 (# # 1.6 µs), transient spectra recorded withthe O2-saturated solution (see Figure 2) showed clear evidencefor the formation of a new transient product ("max # 370 nm),which grows in over a similar time scale as the silylene decayand is then consumed on a time scale of several microseconds(# # 3.3 µs). While we clearly have insufficient data for areliable assignment for the species to be made at the presenttime, the lifetime seems rather short and the absorptionmaximum too far to the red for it to be due to 1,1-diphenyl-dioxasilirane (18; R ) Ph); the similarity in the spectrum tothose of the complexes of SiPh2 with the other O-donors studiedherein suggests the triplet silanone oxide may be the most likelycandidate for the species. It should be noted that the spectrumof the species is strikingly similar to the matrix spectrum ofthe SiMes2-O2 adduct reported by Ando and co-workers.37 Itshould also be noted that the fact that the species is significantlylonger-lived than free SiPh2 under the same conditions indicatesthat it is formed irreversibly.Nothing but SiMe2 (# # 100 ns), Si2Me4 (# # 250 ns), and

the long-lived photolysis coproduct ("max ) 280 nm) could bedetected in experiments with 1 in O2-saturated hexane, sounfortunately we cannot address the mechanistic details of thereaction of SiMe2 with O2 in fluid solution. We can, however,

provide an estimate of kO2 ) (2.9 ( 0.4) " 108 M-1 s-1 for therate constant for reaction of Si2Me4 with O2 under theseconditions, based on the lifetimes of the species in air- and O2-saturated solution. The value compares favorably with thosethat have been measured for other transient disilenes in solution43

and is a factor of about 5 larger than that for the correspondingreaction of the germanium analogue, Ge2Me4.44

Further studies of the reactions of arylsilylenes with molecularoxygen in hydrocarbon solvents are in progress.

Reactions of SiPh2 and SiMe2 with Group 14 Hydrides

R3MH (M ) Si, Ge, Sn). The Si-H insertion reaction ofsilylenes with hydridosilanes is among the best known ofsilylene reactions and has been particularly extensively studiedin the gas phase with several simple silylene derivatives suchas SiH2, SiMeH, SiPhH, SiClH, and SiMe2 as a function ofmethyl substitution in the silane (MenSiH4-n; n ) 0-3).4,5 Theaccepted mechanism involves the initial formation of a weaklybound H-bonded silylene(acceptor)-hydridosilane(donor) com-plex, which rearranges to yield the final insertion product (eq11); the latter is thought to be the rate-determining step in thereactions of the substituted silylenes that have been studied.4,45

The effects of substituents on the high-pressure-limiting rateconstants for these reactions in the gas phase vary somewhatdepending on the degree of methyl substitution in the silane,but with trimethylsilane (Me3SiH) vary in the order PhSiH !MeSiH ! SiH2 . SiMe2 > ClSiH.5 The effect has beenrationalized as the result of destabilization of both the intermedi-ate complex and the barrier for the second step, relative to thesituation in the corresponding reaction of the parent molecule(SiH2), with the effect being much greater for SiMe2 than forMeSiH and PhSiH.45 Ge-H insertions have also receivedattention in gas-phase studies, though much less extensively thanis the case for Si-H insertions. Surprisingly, SiMe2 has recentlybeen found to react close to 10 times more slowly with Me2-GeH2 than with Me2SiH2, which runs counter to what might beexpected on the basis of the relative strengths of the M-H bondsin the two substrates.5

The rate constant determined in the present work for reactionof SiMe2 with Et3SiH in hexane (kEt3SiH ) (3.6 ( 0.3) " 109

M-1 s-1) is the same as the values reported previously incyclohexane within experimental error,8,9 as should be expectedgiven that it is well below the diffusion-controlled limit, and isalso in good agreement with the value reported for reaction ofthe same silylene with Me3SiH in the gas phase (k ) 2.7 " 109

M-1 s-1).46 The value is also quite similar to the value foundfor SiPh2+ Et3SiH in hexane,6,16 indicating once again the closesimilarities in the reactivities of SiMe2 and SiPh2. The twosilylenes are ca. 40 times more reactive toward Si-H insertionthan SiMes2 in cyclohexane under similar conditions,11 indicat-ing that steric effects on the Si-H insertion reaction aremoderate and of a similar magnitude to those observed in thereaction with MeOH.16

Silylene insertions into Ge-H and Sn-H bonds have beenmuch less widely studied than Si-H insertions, but at least one

(42) Battino, R.; Rettich, T. R.; Tominaga, T. J. Phys. Chem. Ref. Data1983, 12, 163.(43) Morkin, T. L.; Owens, T. R.; Leigh, W. J. In The Chemistry of

Organic Silicon Compounds, Vol. 3; Rappoport, Z., Apeloig, Y., Eds.; JohnWiley and Sons: New York, 2001; pp 949-1026.

(44) Leigh, W. J.; Lollmahomed, F.; Harrington, C. R. Organometallics2006, 25, 2055.(45) Becerra, R.; Carpenter, I. W.; Gordon, M. S.; Roskop, L.; Walsh,

R. Phys. Chem. Chem. Phys. 2007, 9, 2121.(46) Baggott, J. E.; Blitz, M. A.; Frey, H. M.; Walsh, R. J. Am. Chem.

Soc. 1990, 112, 8337.

Figure 2. Transient absorption spectra recorded 0-48 ns (-0-),0.54-0.62 µs (-O-), and 2.14-2.22 µs (-4-) after the laser pulse,by laser flash photolysis of a 0.09 mM solution of 9 in oxygen-saturated hexane at 25 °C; the inset shows transient growth/decayprofiles recorded at 520 (O), 460 (4), and 350 nm (0).


example of each process is known for transient silylenes suchas SiMe247 and SiMePh19 in solution. The expected M-Hinsertion products from SiPh2 (20 and 21; eq 12) are obtainedalong with 10 in excellent yields upon photolysis of 9 in C6D12in the presence of Et3GeH and n-Bu3SnH, respectively (see eq12). Figure 3a shows the plots of kdecay versus [Et3GeH] forSiMe2 and SiPh2 in hexane at 25 °C, while Figure 3b showstime-resolved absorption spectra recorded by flash photolysisof a solution of 9 containing 1.7 mM Et3GeH, where the lifetimeof SiPh2 is reduced to # # 300 ns and the formation of Si2Ph4is strongly quenched. The spectra illustrate the absence ofdetectable intermediates in the primary reaction of SiPh2 withEt3GeH; similar results were obtained with Bu3SnH.

Perhaps surprisingly given the systematic decrease in M-Hbond strengths throughout the series M ) Si > Ge > Sn,48 therate constants for the reactions of SiPh2 with the three metallylhydrides are all quite similar, with those for the Si-H andGe-H insertions being essentially identical and less than a factorof 2 lower than that for the Sn-H insertion. A somewhat largerspread in reactivity is observed for SiMe2, with the reactionwith Et3GeH proceeding slightly slower than with Et3SiH, andthat with Bu3SnH ca. 5 times faster. The difference in rateconstants for reaction of SiMe2 with the germane and silane issmall, but is in the same direction that has been observed forthe reactions of Me2GeH2 and Me2SiH2 with SiMe2 in the gasphase,5 as mentioned above. The absence of detectable transientproducts in laser photolysis experiments with SiPh2 in thepresence of these substrates verifies, for the Sn-H insertion inparticular, that the reaction does not proceed via H atomabstraction, as the tributylstannyl radical should be readilydetectable if it was formed as a discrete intermediate.49,50

Reactions of SiPh2 and SiMe2 with Alkenes and Alkynes.

The reactions of silylenes with C-C unsaturated compoundshave had a long and fascinating history, which is welldocumented in early reviews.1 The corresponding three-membered ring compoundsssiliranes from alkenes and silirenesfrom alkynessare well known to be the primary products ofthese reactions, which proceeds stereospecifically in the caseof addition to alkenes.20,51-53 The stereochemistry of the additionto alkenes was in fact first established with SiMe2 and SiPh2,employing the photolysis of 1 and 4 to generate the two speciesin the presence of high concentrations of cis- or trans-2-buteneand trapping the resulting siliranes, which are often ratherunstable, as their methanolysis products.20,51

The kinetics of the reactions of both SiMe2 and SiMes2 withalkenes were examined in the earlier flash photolysis studiesof these species in cyclohexane solution, with values of 7.3 "109 and 5.9 " 109 M-1 s-1 being reported for SiMe2 with1-hexene and trimethylsilylethylene, respectively,8 and k ) 2.8" 106 M-1 s-1 for SiMes2 with cyclohexene.11 Rate constantsfor the reactions of SiMe2 and SiPh2 with cyclohexene and theterminal alkene 3,3-dimethyl-1-pentene (DMP) have beenmeasured in the present work and again reveal close similaritiesbetween the reactivities of the two species. Though we havenot carried out product studies in either case, Tortorelli et al.showed that SiMe2 and SiPh2 both react with cyclohexene by1,2-addition, the latter yielding 22 (see Chart 1).20 Steric effectsare felt considerably more strongly in this reaction than inMeOH or Et3SiH insertions, as shown by the ca. 2000-fold lowerrate constant for reaction of SiMes2 with cyclohexene11 com-pared to that for SiPh2 (k ) 7.9 " 109 M-1 s-1). The differenceis of a similar magnitude to that observed for the complexationreaction with MeOTMS; in both cases, it is impossible to avoidsome degree of nonbonded interaction between the substituentson the silylene and the substrate in the transition state for therate-determining step of the reaction, and hence the sensitivityof the rate constant to steric bulk associated with the silylenesubstituents is greater.

The rate constants for reaction of the three silylenes with 2,3-dimethyl-1,3-butadiene (DMB) and isoprene are 2-3 timeslarger than those for reaction with the alkenes, but exhibit asimilar trend with silylene structure, decreasing in the order

(47) Appler, H.; Neumann, W. P. J. Organomet. Chem. 1986, 314, 247.(48) Luo, Y.-R. In Handbook of Bond Dissociation Energies in Organic

Compounds; CRC Press LLC: Boca Raton, 2003; p 287.(49) Chatgilialoglu, C.; Ingold, K. U.; Lusztyk, J.; Nazran, A. S.; Scaiano,

J. C. Organometallics 1983, 2, 1332.(50) Shaw, W. J.; Kandandarachchi, P.; Franz, J. A.; Autrey, T.

Organometallics 2004, 23, 2080.

(51) Tortorelli, V. J.; Jones, M., Jr. J. Am. Chem. Soc. 1980, 102, 1425.(52) Zhang, S.; Wagenseller, P. E.; Conlin, R. T. J. Am. Chem. Soc.

1991, 113, 4278.(53) Pae, D. H.; Xiao, M.; Chiang, M. Y.; Gaspar, P. P. J. Am. Chem.

Soc. 1991, 113, 1281.

Figure 3. (a) Plots of the pseudo-first-order decay rate constants (kdecay) for SiMe2 (0) and SiPh2 (O) vs [Et3GeH] in hexane solution at25 °C. The solid lines are the linear least-squares fits of the data to eq 6. (b) Transient absorption spectra recorded 16-80 ns (-0-) and0.86-0.98 µs (-O-) after the laser pulse, by laser flash photolysis of a 0.09 mM solution of 9 in deoxygenated hexane containing 1.7 mMEt3GeH. The inset shows transient growth/decay profiles recorded at 460 and 530 nm.


kSiMe2 ! kSiPh2 ! 1600kSiMes2. The reaction is thought to proceedanalogously to that with alkenes, forming the corresponding2-vinyl-1-silirane derivative as the main primary product, whichundergoes rapid secondary rearrangement to the isomeric1-silacyclopent-3-ene and (under photochemical conditions)other (acyclic) isomers.1 The evidence for vinylsilirane formationhas come largely from trapping experiments, but was put on afirm footing by Zhang and Conlin’s successful isolation ofseveral vinylsilirane derivatives from reaction of SiMes2 withaliphatic dienes.56 The isomeric 1-silacyclopent-3-ene is appar-ently always formed in the reactions of silylenes with dienes,but has been suggested to arise mainly via secondary thermaland/or photochemical rearrangement of the vinylsilirane asopposed to a direct (4+1) cycloaddition pathway.1,18

The reaction of SiPh2 with DMB was first studied by Jonesand co-workers, who isolated silacyclopentene 24 and the formalene-type product 25 from photolysis of 4 in the presence of thediene (eq 13).20 Both compounds were suggested to arise from“rapid isomerization” of vinylsilirane 23, citing the earlierconclusions of Ishikawa, Ohi, and Kumada regarding the originsof the analogous products from photolysis of the SiMePhprecursor 3 in the presence of DMB under similar conditions.54,55

In light of this and a recent study of our own of thephotochemistry of 24,57 we were surprised to discover thatvinylsilirane 23 is easily detectable by NMR spectroscopy uponphotolysis of 9 in C6D12 in the presence of DMB, where it isformed in a chemical yield of ca. 87% along with minor amountsof silacyclopentene 24 (eq 14). The structural assignment isbased on comparison of the 1H and 29Si NMR spectra of thephotolyzed mixture (see Supporting Information) to the datareported by Zhang and Conlin for the corresponding vinylsiliraneobtained from reaction of DMB with SiMes2.56 As expected,the compound shows marked thermal instability, rearrangingwithin 48 h (in the dark) to yield 24. Isomer 25 could not bedetected as a coproduct in this experiment, indicating that thequantum yield for its formation by secondary photolysis of 23is substantially lower than that for formation of 24 by the sameroute. The concentration versus time plot for product 24 showsgood linearity over the 0-30% conversion range in 9 (seeSupporting Information), which would seem to be inconsistentwith it being formed exclusiVely via secondary rearrangementof vinylsilirane 23, but rather also via direct (4+1) cycloaddition.However, it should be pointed out that the lowest conversionprobed in our steady-state photolysis experiment was ca. 5%,and the ratio of 24:23 present appeared to be significantly lowerthan its value at ca. 10% conversion. It thus seems more likelythat 24 is formed mainly via secondary rearrangement of 23,

which is consistent with the results of Bobbitt and Gaspar forthe reaction of SiPh2 with 1,3-butadiene in the presence ofacetone.1,18

While numerous examples illustrating the course of thereaction of silylenes with alkynes have been reported, to ourknowledge none have specifically addressed SiPh2. We thusphotolyzed 9 in the presence of bis(trimethylsilyl)acetylene(BTMSE) to verify the course of the reaction expected on thebasis of the reported studies of the reaction of SiMe258 andSiMes259 with this alkyne. Indeed, a single major (SiPh2-derived)product was formed in the reaction and was assigned to silirene26 on the basis of the 1H and 29Si NMR spectra of the crudephotolysate (see eq 15). The 29Si spectrum showed, in additionto the resonances due to 9 and the coproduct 10, resonances at$ -11.37 and-114.46 due to the SiMe3 and ring-silicon atoms,respectively, in 26; the latter is in close correspondence withthe analogous data reported for the silirenes derived fromreaction of the same alkyne with SiMe258 and SiMes2.59 As withthe other C-C unsaturated compounds studied, near diffusion-controlled quenching was observed for both SiMe2 and SiPh2in hexane solutions of 1 and 9 containing submillimolarconcentrations of BTMSE, leading to absolute rate constantsof kBTMSE ) 1.38 " 1010 and 7.6 " 109 M-1 s-1 for SiMe2 andSiPh2, respectively.

Transient absorption spectra recorded with hexane solutionsof 9 containing 0.4-1.0 mM DMP, DMB, and BTMSE,conditions under which the lifetime of SiPh2 is reduced to # <200 ns, showed absorptions centered at ca. 280 nm that did notdecay during the maximum time window (of 0.9 s) that can bemonitored with our system (see Supporting Information). Thespectra and extended lifetimes of these species are consistentwith their assignment to the corresponding three-membered-ring compounds. The UV/vis spectrum of the product formedfrom DMB (see Supporting Information) is identical to thatreported by us previously for the long-lived product detectedin laser photolysis experiments with silacyclopentene24, whichwas assigned to vinylsilirane 23.57 No spectroscopic evidencefor intermediate complexes could be obtained for any of thesereactions.Reactions of SiPh2 and SiMe2 with CCl4. The reactions of

silylenes with halocarbons have also been of considerableinterest.17,47,60-70 The reaction varies in overall course depending

(54) Ishikawa, M.; Ohi, F.; Kumada, M. J. Organomet. Chem. 1975,86, C23.(55) Ishikawa, M.; Nakagawa, K.-I.; Enokida, R.; Kumada, M. J.

Organomet. Chem. 1980, 201, 151.(56) Zhang, S.; Conlin, R. T. J. Am. Chem. Soc. 1991, 113, 4272.(57) Leigh, W. J.; Huck, L. A.; Held, E.; Harrington, C. R.Silicon Chem.

2005, 3, 139.

(58) Seyferth, D.; Annarelli, D. C.; Vick, S. C. J. Am. Chem. Soc. 1976,98, 6382.(59) Ishikawa, M.; Nishimura, K.; Sugisawa, H.; Kumada, M. J.

Organomet. Chem. 1980, 194, 147.(60) Ishikawa, M.; Nakagawa, K.-I.; Katayama, S.; Kumada, M. J.

Organomet. Chem. 1981, 216, C48.(61) Ishikawa, M.; Nakagawa, K.-I.; Katayama, S.; Kumada, M. J. Am.

Chem. Soc. 1981, 103, 4170.


on the silylene and the halocarbon; for example, transientsilylenes such as SiMe2 undergo C-Cl insertion with monochlo-roalkanes60,71 and halogen atom abstraction with CCl4, the latterleading ultimately to the formation of the correspondingdichlorosilane and other radical-derived products.62-65,69 Thelatter is precisely the behavior exhibited by SiPh2, as shown bythe results of steady-state photolysis of a solution of 9 in C6D12containing 0.2 M CCl4 (eq 16). The reaction affords dichlo-rodiphenylsilane (27; 78%) as the main SiPh2-derived product,in addition to 10 (89%), CHCl3 (25%), and hexachloroethane,with the latter being detected (but not quantified) by GC/MS;we did not determine whether chloroform-d was formed alongwith the protiated isotopomer in the reaction. At least oneadditional product was detected by 1H NMR spectroscopy,which indicated it to be formed in ca. 25% yield relative toconsumed 9. This could not be confirmed by GC/MS, however.The NMR spectrum suggests that the product retains the ringstructure of the starting material, and thus it seems most likelythat it is a radical-derived product resulting from reaction of 9with trichloromethyl radicals. Chlorodiphenylsilane-d (Ph2Si-(D)Cl) was ruled out as a possible product by spiking the crudesample with a small quantity of an authentic sample of theprotiated isotopomer.

The reaction is thought to proceed in general via theintermediacy of a silylene-halocarbon Lewis acid-base com-plex,60,66,69,70 whose partitioning between halogen atom abstrac-tion and other pathways depends primarily on the nature of thehalocarbon; Cl abstraction is known to be very strongly favoredin the case of CCl4.47,63,66 No absolute rate constants have yetbeen reported for a silylene-halocarbon reaction, although

Taraban and co-workers have studied the photochemistry of abenzosilanorbornadiene-type precursor to SiMe2 (28) in thepresence of CCl4 by flash photolysis and CIDNP methods andassigned a relatively long-lived transient they observed by flashphotolysis in hexane containing 2 M CCl4 to the SiMe2-CCl4complex.69 The free silylene cannot be detected directly in laserphotolysis experiments with this precursor,72,73 and so an indirectmethod was employed to estimate a value of k # 106 M-1 s-1

for the rate constant for complexation of the (singlet) silylenewith CCl4.69

Addition of CCl4 to hexane solutions of 1 and 9 led in bothcases to linear dependences of the silylene decay rate constantson CCl4 concentration (see Figure 4a) and concomitant quench-ing of the disilene signal intensities, as was observed with theother scavengers. Analysis of the kdecay-concentration dataaccording to eq 6 afforded absolute rate constants of kCCl4 )3.4 " 109 and 1.37 " 109 M-1 s-1 for SiMe2 and SiPh2,respectively. Transient absorption spectra recorded by laserphotolysis of 9 in the presence of 4 mM CCl4, where the lifetimeof SiPh2 was reduced to # # 225 ns, showed the characteristicabsorption bands due to SiPh2 and the long-lived residualabsorber ("max) 460 nm), both of which were formed promptlywith the laser pulse (see Figure 4b). The lifetime of the latterspecies (# # 1.4 µs) was roughly 100-fold shorter than its valuein neat hexane; this indicates a quenching rate constant of kQ# 2 " 108 M-1 s-1, which is again consistent with the silene(11) assignment.32 Interestingly, the signals in the 330-350 nmregion of the spectrum of 9 in the presence of 4 mM CCl4exhibited somewhat longer lifetimes compared to those at 300and 530 nm (see inset, Figure 4b), suggesting the presence of

(62) Kira, M.; Sakamoto, K.; Sakurai, H. J. Am. Chem. Soc. 1983, 105,7469.(63) Nakao, R.; Oka, K.; Dohmaru, T.; Nagata, Y.; Fukumoto, T. J.

Chem. Soc., Chem. Commun. 1985, 766.(64) Kolesnikov, S. P.; Egorov, M. P.; Galminas, A. M.; Ezhova, M.

B.; Nefedov, O. M.; Leshina, T. V.; Taraban, M. B.; Kruppa, A. I.;Maryasova, V. I. J. Organomet. Chem. 1990, 391, C1.(65) Lehnig, M.; Klaukien, H.; Reininghaus, F. Ber. Bunsenges. Phys.

Chem. 1990, 94, 1411.(66) Oka, K.; Nakao, R. J. Organomet. Chem. 1990, 390, 7.(67) Moser, D. F.; Bosse, T.; Olson, J.; Moser, J. L.; Guzei, I. A.; West,

R. J. Am. Chem. Soc. 2002, 124, 4186.(68) Su, M.-D. J. Am. Chem. Soc. 2003, 125, 1714.(69) Taraban, M. B.; Volkova, O. S.; Plyusnin, V. F.; Kruppa, A. I.;

Leshina, T. V.; Egorov, M. P.; Nefedov, O. M. J. Phys. Chem. A 2003,107, 40962.(70) Becerra, R.; Cannady, J. P.; Walsh, R. J. Phys. Chem. A 2006, 110,

6680.(71) Ishida, S.; Iwamoto, T.; Kabuto, C.; Kira, M. Chem. Lett. 2001,

2001, 1102.

(72) Hawari, J. A.; Griller, D. Organometallics 1984, 3, 1123.(73) Taraban, M. B.; Volkova, O. S.; Kruppa, A. I.; Plyusnin, V. F.;

Grivin, V. P.; Ivanov, Y. V.; Leshina, T. V.; Egorov, M. P.; Nefedov, O.M. J. Organomet. Chem. 1998, 566, 73.

Figure 4. (a) Plots of the pseudo-first-order decay rate constants (kdecay) for SiMe2 (-0-) and SiPh2 (-O-) vs [CCl4] in hexane solution at25 °C. The solid lines are the linear least-squares fits of the data to eq 6. (b) Transient absorption spectra recorded 48-80 ns (-0-),0.32-0.43 (-O-), and 0.88-0.99 µs (-4-) after the laser pulse, by laser flash photolysis of a 0.09 mM solution of 9 in deoxygenated hexanecontaining 4 mM CCl4. The inset shows transient growth/decay profiles recorded at 330, 460, and 530 nm.


a transient product that absorbs in this wavelength range. Theidentity of this species is uncertain, but we consider the mostlikely possibility to be the chlorodiphenylsilyl radical (Ph2SiCl),which the product studies suggest is formed as the primaryproduct of the reaction of the silylene with the halocarbon. Thelifetime of the species shows a similar (but slightly lower)sensitivity to CCl4 concentration than SiPh2; a plot of kdecayvalues at 330 nm versus [CCl4] was linear, affording a rateconstant of kQ ) (8 ( 1) " 108 M-1 s-1. The value is similarto those reported for the (Cl abstraction) reactions of otherarylsilyl radicals with CCl4.74

Transient spectra recorded with a solution of 1 containing0.6 mM CCl4 showed only the characteristic absorptions dueto SiMe2 (# # 180 ns), Si2Me4 (peak intensity ca. 25% of thatin pure hexane; # # 680 ns), and the long-lived photoproductthat is characteristic of all the experiments carried out with thisprecursor; no new transient products or intermediates could bedetected in the experiment (see Supporting Information). Thelifetime of Si2Me4 under these conditions provides an estimateof k # 2.4 " 109 M-1 s-1 for the rate constant for reaction ofthe disilene with CCl4, which is thought to proceed via initialCl atom abstraction.75,76 This is 10 orders of magnitude higherthan the value reported by Kira and co-workers for reaction ofCCl4 with a stable tetrasilyldisilene in solution under similarconditions.75 The value for Si2Me4 can also be compared to thatfor reaction of the germanium analogue, Ge2Me4, with CCl4 inhexane at 25 °C, k ) 2.3 " 107 M-1 s-1.44 The difference inrates for the two tetramethyldimetallenes is in order-of-magni-tude agreement with the recent calculations of Su on the ener-getics of Cl atom abstraction from CCl4 by these compounds.76

Summary and Conclusions

Absolute rate constants have been determined for the reactionsof the transient silylenes SiMe2 and SiPh2 with a wide varietyof representative reagents, under a common set of conditions.The reactions that have been studied are all well known, andmany of the associated mechanistic details have already beenreasonably well established with the aid of product studies,competition experiments, and low-temperature (matrix) spec-troscopic studies. Gas-phase kinetics and theoretical studies ofthe parent molecule and the few other small molecules that lendthemselves readily to work of this type have also played aparticularly significant role in developing our understanding ofsilylene reactivity. In particular, this work provides a quantitativelink between the behavior of the parent molecule, for whichkinetic studies are limited to the gas phase, and simplesubstituted derivatives such as SiMe2, which can be studied inboth the gas phase and solution.The present work has been directed at further developing the

links between these various efforts and exploring some of thefiner mechanistic details that are associated with silylenechemistry. We have shown that substitution of methyl withphenyl groups generally leads to almost no change in intrinsicreactivity toward a wide variety of representative substrates;where differences do exist, they are invariably small, with SiPh2being only slightly less reactive than SiMe2. Similar trends areevident in the relative reactivities of Si(H)Ph and Si(H)Metoward the methylsilanes (MenSiH4-n; n ) 0-3) in the gasphase, as mentioned earlier in the paper.5 We have fewer dataavailable at present to address the differences in reactivity

between SiMe2- and SiPh2-derived reaction intermediates, asonly a few of the reactions studied in the present work proceedvia intermediates that build up in sufficient concentrations forus to detect them. Nevertheless, the few examples for whichthis is the case indicate that the differences here are also small.Those derived from SiPh2 tend to be longer-lived, absorb atlonger wavelengths, and exhibit more intense UV/vis spectrathan their SiMe2-derived counterparts, all distinct advantagesin studies of this type. Though it has not been specificallyquantified, dimerization appears to be somewhat slower in thecase of SiPh2, which imparts scavenging reactions with intrinsi-cally greater efficiency; this makes kinetic analyses more reliableat low substrate concentrations and provides better sensitivityin experiments directed at the detection and characterization ofreaction intermediates that are involved in some cases. Thephenylated system has the additional advantage of being readilyamenable to the introduction of polar ring substituents, whichwill undoubtedly prove to be very useful in future, more detailedmechanistic studies of silylene reactivity.Wherever possible, we have made a point of comparing our

kinetic data for SiPh2 to previously reported rate constants forthe corresponding reactions of the sterically hindered diarylsi-lylene, SiMes2,11 in the interest of developing kinetic bench-marks describing the sensitivity of various silylene reaction typesto steric effects. A convenient measure of this sensitivity is therelative rate constants for reaction of SiMes2 and SiPh2 with agiven substrate, subject to the proviso that these rate ratios mayalso contain contributions to the rate constants from theelectronic effects of the mesityl methyl groups; there are six ofthem, so this effect may not be insignificant. The list of reactionsfor which direct comparisons can be made is somewhat limitedat present, but it is nevertheless already clear that trends ofvaluable predictive power will emerge from further comparisonsof this type. The seven substrates for which we have data divideinto three groups: those with little apparent sensitivity to stericeffects (acetone, O2; kSiMes2/kSiPh2 > 0.1), those with moderatesensitivity (MeOH, Et3SiH; 0.01 < kSiMes2/kSiPh2 < 0.1), andthose with high sensitivity (cyclohexene, DMB, MeOTMS;kSiMes2/kSiPh2 e 10-3). These trends make good sense whenconsidered against the expected structures of the transition statesfor the rate-determining steps in the various reactions and thedegree of nonbonded interactions that must develop betweenthe substrate and the silylene substituents. This is illustratedpictorially in Figure 5 for the reactions with acetone, O2, MeOH,

(74) Chatgilialoglu, C. Chem. ReV. 1995, 95, 1229.(75) Kira, M.; Ishima, T.; Iwamoto, T.; Ichinohe, M. J. Am. Chem. Soc.

2001, 123, 1676.(76) Su, M.-D. J. Phys. Chem. A 2004, 108, 823.

Figure 5. Representations of the transition states for the rate-determining steps in the reactions of silylenes with varioussubstrates, illustrating the variation in silylene-substrate stericinteractions with reaction type. Also shown are the rate constantratios kSiMes2/kSiPh2 for each of the systems pictured, which providesa quantitative description of the effect.


Et3SiH, DMB, and cyclohexene. The reader is directed to theextensive body of theoretical work that has been done on mostof these reactions in the case of the parent molecule, for moreaccurate depictions of the transition-state structures.40,41,45,77-80

A few of the systems studied in this and the preceding paperhave afforded particularly rich mechanistic information as aresult of the successful detection of intermediates formed asthe primary products of the silylene-substrate reactions. Theseinclude the reactions of SiPh2 with methanol,16 methoxytrim-ethylsilane, CCl4, and oxygen. In all four cases, intriguingpreliminary results have been obtained, underlining the fact thatmuch remains to be learned of the mechanistic details of manysilylene reactions. Future work from our laboratory will addressthese reactions, and others, in greater detail.

Experimental Section

1H and 13C NMR spectra were recorded on Bruker AV200 orAV600 spectrometers in deuterated chloroform and were referencedto the solvent residual proton and 13C signals, respectively, while29Si spectra were recorded on the AV600 using the HMBC pulsesequence and referenced to an external solution of tetramethylsilane.GC/MS analyses were determined on a Varian Saturn 2200 GC/MS/MS system equipped with a VF-5ms capillary column (30 m" 0.25 mm; 0.25 µm; Varian, Inc.). High-resolution electron massspectra and exact masses were determined on a Micromass TofSpec2E mass spectrometer using electron impact ionization (70 eV).Infrared spectra were recorded as thin films on potassium bromideplates using a Bio-Rad FTS-40 FTIR spectrometer. Columnchromatography was carried out using a 3 " 60 cm column usingsilica gel 60 (230-400 mesh; Silicycle).1,1,3,3-Tetramethyl-2,2-diphenyl-1,2,3-trisilacyclohexane (9) was

synthesized according to the method described in the precedingpaper,16 while dodecamethylcyclohexasilane (1) was used asreceived from Sigma-Aldrich. Methoxytrimethylsilane was syn-thesized by reaction of methanol with excess hexamethyldisila-zane and purified by distillation (bp 56-58 °C).81 All the otherscavengers investigated in this work were obtained from commercialsources in the highest purity available. Triethylsilane (Et3SiH),triethylgermane (Et3GeH), and tri-n-butylstannane (Bu3SnH) werestirred at room temperature for 18 h over lithium aluminum hydrideand then either distilled at atmospheric pressure (Et3SiH) or undermild vacuum (Bu3SnH) or passed through a short silica gel columnin the case of Et3GeH. CCl4 was refluxed over phosphorus pentoxideand distilled. 4,4-Dimethyl-1-pentene (DMP), isoprene, and 2,3-dimethyl-1,3-butadiene (DMB) were purified by passage of the neatliquids through a silica gel column. 3,3-Dimethyl-1-butyne (TBE)and cyclohexene were distilled. Bis(trimethylsilyl)acetylene (Gelest),acetone (Caledon Reagent), and glacial acetic acid (Sigma-Aldrich)were used as received from the suppliers. Hexanes (EMD Omni-Solv), diethyl ether (Caledon Reagent), and tetrahydrofuran (Cale-don Reagent) were dried by passage through activated aluminaunder nitrogen using a Solv-Tek solvent purification system (Solv-Tek, Inc). Deuterated solvents were used as received fromCambridge Isotope Laboratories.Triethyl(diphenylsilyl)germane (20). In an oven-dried 15 mL

two-necked round-bottom flask fitted with condenser, argon inlet,and magnetic stir bar was placed diisopropylamine (0.17 g, 1.7mmol) in THF (2 mL) under argon, and the solution was cooled to

-78 °C with a dry ice/acetone bath. A 2 M solution of n-butyllithium (0.82 mL, 1.65 mmol) in pentane was added dropwisevia syringe over 20 min, and the mixture was warmed to ca. 0 °Cin an ice bath and then kept at this temperature for 15 min. Themixture was recooled to -78 °C, and a solution of triethylgermane(0.27 g, 1.69 mmol) in THF (1.5 mL) was added dropwise over 1h. The bath was removed, and the contents were taken up in asyringe and then added dropwise over 1 h to a solution ofchlorodiphenylsilane (0.31 g, 1.42 mmol) in THF (2 mL) underargon at -40 °C (acetonitrile-dry ice bath) in an oven-dried 25mL two-necked round-bottom flask fitted with condenser, argoninlet, and magnetic stir bar. The reaction mixture was left to stirovernight, with the temperature rising slowly to room temperature.The mixture was diluted with diethyl ether (5 mL) and filteredthrough a short silica gel column to remove white precipitate andpolar impurities, and the solvent was removed under vacuum toyield a colorless oil (0.27 g). Column chromatography on silicagel with hexanes as eluant afforded triethyl(diphenylsilyl)germane(20; 0.077 g, 16%), which was identified on the basis of thefollowing spectroscopic data: 1H NMR (600 MHz, CDCl3) $ 0.96(q, J ) 7.8 Hz, 6H), 1.05 (t, J ) 7.8 Hz, 9H), 5.08 (s, 1H), 7.37-(m,6H), 7.58 (dd, J ) 1.8, 7.8 Hz, 4H); 13C NMR (150 MHz,CDCl3) $ 9.0, 13.8, 27.5, 30.1, 128.2, 129.1, 135.0, 135.8; 29SiNMR (119 MHz, CDCl3) $ -29.25 (Ph2SiH, JSi-H ) 186.0 Hz);IR (neat) % (cm-1) ) 3068 (m), 3052 (m), 2950 (s), 2928 (s), 2909(s), 2874 (s), 2099 (s), 1486 (m), 1459 (m), 1430 (s), 1104 (m),1014 (m), 791 (s), 754 (s), 722 (s); GC/MS (EI)m/z ) 344(8; M+),317 (17), 316 (18), 315 (65), 314 (32), 313 (61), 311 (47), 289(23), 288 (22), 287 (100), 286 (40), 285 (80), 284 (17), 283 (56),261 (14), 260 (9). 259 (49), 258 (19), 257 (49), 256 (13), 255 (39),253 (9), 212 (16), 184 (16), 183 (83), 181 (26), 151 (31), 149 (25),147 (15), 135 (15), 133 (12), 131 (11), 107 (18), 105 (65), 103(27), 101 (13), 75 (10), 53 (17); exact mass calculated forC18H26Si74Ge 344.1016, found 344.1024.

Tri-n-butyl(diphenylsilyl)stannane (21). The compound wasprepared in identical fashion to 20, from tri-n-butyltin hydride(0.37 g, 1.27 mmol) and chlorodiphenylsilane (0.26 g, 1.19 mmol).Column chromatography of the crude reaction mixture on silicagel with hexanes as eluant afforded tri-n-butyl(diphenylsilyl)-stannane (21; 0.055 g, 10%), which was identified on the basis ofthe following spectroscopic data: 1H NMR (600 MHz, CDCl3) $0.84 (t, J ) 7.8 Hz, 9H), 1.00 (t, J ) 7.8 Hz, 6H), 1.26 (sextet, J) 7.8 Hz, 6H), 1.46 (quint, J ) 7.8 Hz, 6H), 7.36 (m,6H), 5.28 (s,1H), 7.57 (m, 4H); 13C NMR (150 MHz, CDCl3) $ 9.0, 13.8, 27.5,30.1, 128.2, 129.1, 135.0, 135.8; 29Si NMR (119 MHz, CDCl3) $-28.23 (Ph2SiH, JSi-H ) 187.4 Hz); IR (neat) % (cm-1) ) 3068(m), 2958 (s), 2871 (s), 2853 (s), 2099 (s), 1464 (m), 1428 (s),1376 (m), 1102 (s), 787 (s), 714 (s), 698 (s); GC/MS (EI) m/z )417 (10; M+ - n-Bu), 416 (7), 415 (9), 414 (6), 361 (31), 360(17), 359 (31), 358 (16), 357 (16), 309 (16), 307 (21), 306 (20),305 (100), 304 (39), 303 (87), 301 (46), 197 (12), 195 (10), 184(11), 183 (50), 181 (14), 177 (9), 121 (8), 105 (17), 79 (3), 53 (5);exact mass calculated for C24H38SiSn 474.1765, found 474.1794;exact mass calculated for C20H29SnSi (M+ - n-Bu) 417.1060, found417.1043.

Steady-State Photolysis Experiments. In a typical procedure,a solution of 9 (0.039-0.051 M) in cyclohexane-d12 (0.7 mL) wasplaced in a quartz NMR tube and the tube was capped with a rubberseptum. The solution was deoxygenated with dry argon for 15 min,and then the appropriate volumes of trapping agent and dioxane(as internal integration standard) were added with a microlitresyringe to result in concentrations of ca. 0.20 and 0.01 M,respectively. The solution was then photolyzed for 18-24 min (ca.44-60% conversion of 9) in a Rayonet photochemical reactor(Southern New England Ultraviolet Co.) equipped with a merry-go-round and two RPR-2537 lamps, monitoring by 600 MHz 1HNMR spectroscopy in 2 min time intervals throughout the experi-

(77) Heaven, M. W.; Metha, G. F.; Buntine, M. A. J. Phys. Chem. A2001, 105, 1185.(78) Heaven, M. W.; Metha, G. F.; Buntine, M. A. Aust. J. Chem. 2001,

54, 185.(79) Sakai, S. Int. J. Quantum Chem. 1998, 70, 291.(80) Skancke, P. N.; Hrovat, D. A.; Borden, W. T. J. Am. Chem. Soc.

1997, 119, 8012.(81) Schrock, R. R.; Sancho, J.; Pederson, S. F. Inorg. Synth. 1989, 26,

44.


ment. Each mixture was also analyzed by GC/MS before and afterphotolysis.

With most of the trapping agents studied, photolysis led to theformation of two major products. The first was common to all andwas identified as 1,1,2,2-tetramethyl-1,2-disilacyclopentane (10) bycomparison of its 1H NMR and mass spectra to reported data:821H NMR (600 MHz, C6D12) $ 0.07 (s, 12H; SiMe2), 0.67 (t, 4H, J) 6.6 Hz; C3,5H2), 1.67 (quint, 2H, J ) 6.6 Hz; C4H2); GC/MS(EI) m/z ) 158 (42; M+), 143 (43), 130 (41), 117 (33), 116 (44),115 (100), 99 (8), 85 (10), 73 (41), 59 (19), 45 (13), 43 (28). Thesecond major product was unique to each of the trapping agentsstudied and could be assigned to the product of reaction of the trapwith SiPh2. Additional minor product(s) were evident in a few cases.In the majority of cases, the silylene-trapping products were identi-fied by comparison of the 1H NMR and mass spectra to literaturedata or to the corresponding spectra of authentic, independentlyprepared samples. Two of the products (23 and 26) were tentativelyidentified from the 1H NMR spectra of the crude photolysismixtures, on the basis of comparisons to the reported spectra ofclosely related analogues. Chemical yields of the major productsformed were determined from the relative slopes of concentration

versus time plots for the product(s) and 9 in each case. Concentra-tions were calculated from the 1H NMR integrals, using the SiMeproton resonances to characterize 9 and 10 and one of the peaksfor the remaining product(s) to characterize it; C-H resonanceswere used wherever possible. Details are provided in the SupportingInformation.

Laser Flash Photolysis Experiments. Laser flash photolysisexperiments were carried out as described in the preceding paper.16

Rate constants were calculated by linear least-squares analysis ofdecay rate-concentration data. Errors in absolute second-order rateconstants are quoted as twice the standard deviation obtained fromthe least-squares analyses.

Acknowledgment. We thank the Natural Sciences andEngineering Research Council of Canada for financial support.

Supporting Information Available: Representative 1H NMRspectra and concentration versus time plots from steady-statephotolysis experiments, plots of kdecay versus [Q] for quenching ofSiMe2 and SiPh2, and transient absorption spectra from laserphotolysis of 9 and 1 in the presence of scavengers. This materialis available free of charge via the Internet at http://pubs.acs.org.

OM700659D(82) Gusel’nikov, L. E.; Polyakov, Yu. P.; Volnina, E. A.; Nametkin,

N. S. J. Organomet. Chem. 1985, 292, 189.


Documents

Comparison of the Reactivities of Dimethylsilylene (SiMe2 ... · 2) in Solution by Laser Flash Photolysis Methods Andrey G. Moiseev and William J. Leigh* Department of Chemistry,